Personalized advertisement pushing method and system

A technology of advertising and advertising, applied in the Internet field, can solve problems such as the poor performance of the advertising recommendation system, and achieve the effect of improving user experience, improving accuracy, and improving accuracy

Pending Publication Date: 2017-06-06
BEIJING ZHANGKUO MOBILE MEDIA TECH
View PDF6 Cites 77 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0029] The technical problem to be solved by the present invention is to provide a method and system for personalized advertisement push, which is used to solve the problem of poor performance of the existing advertisement recommendation system

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Personalized advertisement pushing method and system
  • Personalized advertisement pushing method and system
  • Personalized advertisement pushing method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0090] Specifically, as figure 1 As shown, from the point of view of the training and use of the advertisement click rate prediction model, in order to improve the accuracy of the advertisement click rate prediction, the advertisement must be recommended to the appropriate user, which requires us to be accurate enough for the user's portrait. To get more accurate user portraits, we must collect more user data (not just display and click data) to train the user portrait model, improve the accuracy of user portraits, and have richer data and precision to describe users, Recommending ads to the most matching users can improve the user experience and thus the accuracy of the CTR prediction model.

[0091] like figure 1 , a method for advertising personalized push, specifically including:

[0092] Step 1) Obtain third-party external data through the crawler, and count the behavior data of a certain user from it; obtain the internal data of the display, click and interactive data ...

Embodiment 2

[0099] More specifically, the technical realization principle of the present invention is as follows:

[0100] Step 1: User portrait

[0101] Data collection: The collection of information about the user's interactive information in the advertisement is added to the advertisement, and the external data is obtained through the crawler program, so as to enrich the data and improve the accuracy and coverage of the portrait.

[0102] Persona Model Selection: Use the collected data to train and select a persona model

[0103] User portrait synchronizes the user information obtained by the user portrait to the advertising platform

[0104] Step 2: Train the CTR prediction model, including:

[0105] Collect historical ad impression and click data

[0106] Data preprocessing: convert data into data samples with accurate user information and labels

[0107] Model training and selection: Train and select a CTR prediction model using preprocessed data

[0108] Step 3: Online verific...

Embodiment 3

[0133] like Figure 2-4 As shown, in one embodiment, the present invention first collects the display, click and interaction data of users in the historical advertising platform, and collects external user behavior data through crawler, uses the data to carry out user portrait, and then uses The user data is merged into the user's display data to train the CTR model. Finally, the CTR model is applied to advertising.

[0134] Among them, it mainly includes the following main steps:

[0135] 1) DMP user portrait:

[0136] 1.1. Data collection:

[0137] The collection of information about the user's interactive information in the advertisement is added to the advertisement provision, and the external data, static information data, and relatively stable information of the user are obtained through the crawler program, as shown in the figure, mainly including demographic attributes, business attributes and other aspects of data. This kind of information is self-labeled. If the ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a personalized advertisement pushing method and system. The method comprises the following steps: 1) obtaining third-party external data through a crawler and calculating the behavior data of a certain user; acquiring the internal data of displaying, clicking, and interactive data of the advertisement of the user in the platform; 2) preprocessing the third-party external data and the internal data and converting the same into a data sample with accurate user information and tags; 3) using the preprocessed data for training, selecting the specific model of a user image, forming a user tag library, synchronizing the obtained user information to an advertisement platform, and storing the obtained user information in a cache tag library by the advertisement platform; 4) collecting the historical advertisement display and click data in the platform, forming a CTR advertisement pushing prediction model, and storing the CTR advertisement putting prediction model in a CTR prediction model library; and 5) pushing advertisements to a certain user accurately based on the user information and the CTR advertisement pushing prediction model.

Description

technical field [0001] The invention belongs to the field of Internet, and relates to a system and method for advertising personalized push. Background technique [0002] In the Internet, such as Baidu, there is a lot of web page information, and the target of the theme advertisement is not a user but a certain type of page. By analogy, each web page type corresponds to a user in the recommender system, and each advertisement corresponds to an item in the recommender system. [0003] The operation process of computational advertising from preliminary investigation to planning, production, delivery, feedback, and then to effect measurement is essentially a process of combining a series of algorithmic models. The purpose of computational advertising is to automatically find the best match between the advertisement, the advertising environment and the audience through a collection of algorithms. The realization of this automated best match is the result of the interaction of ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q30/02
CPCG06Q30/0256G06Q30/0271G06Q30/0277
Inventor 胡云志
Owner BEIJING ZHANGKUO MOBILE MEDIA TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap